Fridays 2:30 - 3:50 p.m.

The LTI colloquium is a series of talks related to language technologies.  The topics include but are not limited to Computational Linguistics, Machine Translation, Speech Recognition and Synthesis, Information Retrieval, Computational Biology, Machine Learning, Text Mining, Knowledge Representation, Computer-Assisted Language Learning, Intelligent Language Tutoring, and Multimodal Computing and Interaction.  All members of the university are welcome to attend the lectures and schedule meetings with the guest speakers.  Click on a date below to view presentation details or to schedule a meeting.  Videos of the lectures are available to the CMU community only.

Fall 2016 - Spring 2017

09-Sep-2016Grace Hui YangDynamic Information Retrieval ModelingVideo
16-Sep-2016Mari OstendorfFinding Information in DisfluenciesVideo
23-Sep-2016James CrowleySituated Interaction with Smart ObjectsVideo
30-Sep-2016Cristian Danescu-Niculescu-MizilLanguage and Social Dynamics

Video Only available to students registered  for 11-700 fall 2016. Andrew ID login. 

07-Oct-2016Jimmy LinThe Academic:  There and Back AgainVideo
14-Oct-2016Dipanjan Das Two Case Studies in Semantic InferenceVideo Not Available
21-Oct-2016Mid-Semester Break (No Colloquium)
28-Oct-2016Margaret Mitchell From Naming Concrete Objects to Sharing Abstract Thought:  Vision-to-Language Begins to Grow UpVideo
04-Nov-2016Jordan Boyd-GraberOpening up the Black Box: Interactive Machine Learning for Understanding Large Document Collections, Characterizing Social Science, and Language-Based GamesVideo
11-Nov-2016Alexander Rush Interpreting, Training, and Distilling Seq2Seq ModelsVideo
18-Nov-2016Shin'ichi SatohSocial Analysis by Using Large-Scale Broadcast Video ArchiveVideo Not Available
25-Nov-2016Thanksgiving Break (No Colloquium)
02-Dec-2016Marcus RohrbachExplain and Answer:  Intelligent Systems Which Can Communicate About What They SeeVideo
09-Dec-2016  Jacob Eisenstein Social Networks, Social MeaningVideo

Spring 2017:  Doherty Hall, Room 2315

20-Jan-2017Burr SettlesDuolingo:  Improving Language Learning and Assessment with DataVideo 
27-Jan-2017Jason CorsoSparse Modeling in Deep and Cross-Modal EmbeddingsVideo
03-Feb-2017Achim Rettinger

Towards Breaking the Language and Modality Barrier:  Learning Cross-lingual Cross-modal Semantic Representations


10-Feb-2017Adam BergerWhat Healthcare Needs from NLPVideo
17-Feb-2017Taylor Berg-Kirkpatrick  Modeling Continuous Human Artifacts:  Music, Prosody, and ShakespeareVideo
24-Feb-2016 LTI Open House (No Colloquium)                   
03-Mar-2017Jianfeng GaoThe Development of an Open-Domain Neural Dialogue SystemVideo
10-Mar-2017Mid-Semester Break (No Colloquium)
17-Mar-2017Spring Break (No Colloquium)
24-Mar-2017Jason EisnerTractable Deep Models of Sequential StructureVideo
31-Mar-2017Alan W. Black & Lori Levin"What is your purpose here?" 
07-Apr-2017Matthias GrabmairSearch, Read, Argue and Predict: An Introduction to Artificial Intelligence & LawVideo
14-Apr-2017Tadas BaltrušaitisUnderstanding and Interpreting Nonverbal LanguageVideo
CMU Spring Carnival  (No Colloquium)
28-Apr-2017Julia Hirschberg Detecting Sentiment and Emotion toward Disasters in Low Resource LanguagesVideo

Policy for Enrolled Students

LTI Colloquium, 11-700, Spring 2017; Alex Hauptmann, Course Instructor; Junwei Liang, Teaching Assistant

ATTENDANCE  You are required to attend all of the lectures.  Absences will be excused only if you are out of town, sick, or under extreme circumstances.  Generally speaking, an approaching paper deadline or "being really busy" is not an acceptable reason to skip a lecture.  We expect that most students may miss at most one lecture during the whole semester.  Frequent unexcused absences will require you to make up for the missed lectures with a critique based on one lecturer's published research.  If you must miss a lecture, please contact the course TA in advance of your absence.

The colloquium runs from 2:30 p.m. until the end of the Q&A period following the presentation.  It is not acceptable to depart when the speaker finishes the last slide.  Arriving late is impolite and shows poor time-management skills.  If you have a class immediately prior to the colloquium, please try to arrive on time anyway.

COURSE CREDIT  To receive credit for the course, students must attend all lectures and successfully complete an on-line quiz made available at the close of each talk.  To determine attendance and pass/fail grades, at the end of each lecture we will reveal a password to  You will only have until midnight of the same day to access the website, confirm your presence in the lecture hall, and answer a few quiz questions about the content of the lecture.  Students who miss too many lectures or with low grades on the quizzes will be required to submit an eight-page critique of one the lecturer’s research publications at the end of the semester to receive a passing grade. This report should NOT be about the student's own research, even if it is related to the lecturer's work.

LAPTOPS DURING LECTURES  We are able to attract excellent speakers to present talks to our students because we can promise them a large, knowledgeable, and engaged audience.  Although laptops can be used for lecture-related activities like taking notes and looking something up, a large number of open laptops (and, even more egregiously, a large number of eyeballs focusing on them) can demoralize speakers.  So, while we don't want to micro-manage students, we ask that you continue to keep your laptops closed except for brief periods when you need to use them for lecture-related activities.  Please use your best judgment so that we don't need to revisit this issue.